Gabor-Granger Pricing: The Complete Method Guide

Setting the wrong price is one of the most expensive mistakes a brand can make. Price too high and demand collapses. Price too low and you leave margin on the table. Gabor-Granger pricing research gives brand managers and pricing strategists a structured, survey-based way to measure exactly where those thresholds sit. Developed by economists André Gabor and Clive Granger in the 1960s, the method remains one of the most widely used tools in willingness to pay research precisely because it is direct, replicable, and produces outputs that finance and commercial teams can act on immediately.

What is Gabor-Granger pricing?

Gabor-Granger pricing is a survey technique that measures price elasticity by presenting respondents with a specific price and asking a single binary question: would you buy this product at this price? By systematically varying the price shown to different respondents, or to the same respondent across sequential rounds, researchers build a demand curve that shows what percentage of the target audience would purchase at each tested price point.

The method was first described by André Gabor and Clive Granger in their 1964 paper on price sensitivity, and it has been a staple of pricing survey methodology ever since. Its enduring popularity comes from its simplicity. The question format is easy for respondents to understand, the data is straightforward to analyse, and the output maps directly to commercial decisions.

Unlike conjoint analysis, which asks respondents to evaluate products across multiple attributes simultaneously, Gabor-Granger focuses exclusively on price. All other product attributes, features, and brand associations are held constant. This makes it well suited to situations where the product itself is defined and the only remaining question is where to set the price.

How the Gabor-Granger method works (step-by-step)

Running a Gabor-Granger pricing study involves five core steps.

Step 1: Define your price range

Before writing a single survey question, you need a plausible price range. Speak to the commercial team, review competitor pricing, and consider any internal cost floors. The range should span from a price you are confident most people would accept to a price you suspect very few would pay. A well-constructed range neither anchors too low nor alienates respondents with an absurd ceiling.

Step 2: Select your price points

Within that range, choose between 5 and 15 specific price points. Spacing can be even (for example, £5 increments) or weighted toward the commercially relevant zone where you expect demand to be most sensitive. Most practitioners settle on 6 to 10 prices as the sweet spot between granularity and respondent fatigue.

Step 3: Design the survey

Each respondent is shown a price and asked: would you buy [product description] at [price]? The question is binary, yes or no, or sometimes a five-point purchase intent scale that is later collapsed to a binary threshold. In a monadic design, each respondent sees only one price. In a sequential design, the price shown adjusts based on the respondent’s previous answer, higher if they said yes, lower if they said no. Both designs produce equivalent data when samples are large enough. Sequential designs are more efficient but require more careful programming.

Best practice: randomise the starting price across respondents to avoid anchoring effects that could bias the resulting demand curve.

Step 4: Collect responses and build the demand curve

Once fieldwork closes, calculate the percentage of respondents who said they would buy at each price point. Plot price on the x-axis and purchase incidence on the y-axis. This is your demand curve. It should slope downward: fewer people will buy as price increases.

Step 5: Calculate the revenue-optimising price

Multiply each price point by its corresponding purchase incidence percentage. The price that produces the highest product of these two numbers is the revenue-maximising price point. For example, if 60% of respondents say they would buy at £10 and 40% at £15, the revenue index is 600 at £10 and 600 at £15. In this case either price is equivalent on expected revenue, and other factors, such as margin and competitive positioning, should inform the final decision.

“The revenue-optimising price is not always the one with the highest purchase intent. It is the price at which the product of demand and price is greatest.”

Gabor-Granger pricing vs Van Westendorp

The Van Westendorp pricing model (sometimes written Price Sensitivity Meter or PSM) is the other survey-based method most commonly paired with or compared to Gabor-Granger. The two methods answer related but distinct questions, which is why many research briefs include both.

Van Westendorp asks four open-ended questions:

1. At what price would this product be so cheap you would question its quality?
2. At what price would this product be a bargain?
3. At what price would this product be getting expensive, but you might still consider it?
4. At what price would this product be too expensive to consider?

The intersections of the resulting curves define an acceptable price range and an optimal price point. The method requires no predetermined prices, which makes it useful when you are entering a market you know little about.

The practical rule of thumb: use Van Westendorp when you do not know what price range the market will accept, and Gabor-Granger once you have a plausible range and need to pinpoint the revenue-optimal price within it. In many category-entry or re-pricing studies, running both in sequence is the most rigorous approach.

For a deeper look at how an AI panel can stress-test both types of study before fieldwork, see our guide on how to test audience assumptions before fieldwork.

When to use Gabor-Granger pricing research

Gabor-Granger is the right tool in several specific circumstances.

You have an existing product or a line extension. Because the method requires predefined price points, it works best when you already have a sense of the plausible range. For a product extension within an established category, the brand’s existing price architecture provides that anchor.

You need a revenue estimate, not just a preference rank. The demand curve and revenue curve outputs translate directly into financial models. Finance teams and commercial directors can work with these numbers in a way they cannot always work with attitudinal data.

Your product attributes are fixed. Gabor-Granger holds everything constant except price. If you are also testing different feature sets, bundle configurations, or pack sizes, conjoint analysis is more appropriate because it can isolate the value of each attribute independently.

You are not in a highly competitive category. The method does not account for competitor pricing. If your category is characterised by constant competitive price moves, a price sensitivity study that ignores the competitive context may overestimate demand at higher prices. In those situations, a competitive conjoint or a Van Westendorp study framed with competitive context is more robust.

Common mistakes in Gabor-Granger studies

Even experienced researchers make errors in Gabor-Granger studies that undermine the validity of results.

Choosing a price range that is too narrow or too wide

If the range is too narrow, the demand curve will be steep at both ends and you will not capture the full picture of price sensitivity. If it is too wide, respondents at the extreme high end will almost universally say no, which adds noise without insight. Pilot your price range before full launch.

Anchoring respondents with a consistent starting price

Showing every respondent the same starting price anchors their responses. A respondent who sees £20 first may evaluate £30 differently to one who starts at £10. Randomise starting prices across respondents, or use a fully randomised monadic design where each respondent sees only one price.

Using purchase intent scales without a clear cut-off rule

Some practitioners use a 5- or 7-point purchase intent scale instead of a binary yes or no question. This is acceptable, but you must decide before analysis which scale points count as would buy. A common convention is to count only the top 2 boxes. Apply this rule consistently or your demand curve will not be comparable across studies.

Ignoring the difference between claimed and actual purchase behaviour

Stated purchase intent consistently overstates actual behaviour. Respondents in a survey face no financial consequence for saying yes. The standard adjustment is to apply a calibration factor, typically reducing top-box intent by 20 to 40 percent depending on the category and claimed novelty of the product. Failing to apply any adjustment will lead to price recommendations that underperform in market.

Running the study on the wrong audience

Gabor-Granger results are only as good as the sample they come from. If your target buyer is a procurement manager in a B2B context, running the survey with a general consumer panel will produce meaningless price points. Define your target audience tightly before fieldwork begins.

Treating the revenue-maximising price as the only answer

The revenue-maximising price point is one input, not the final answer. Margin, competitive positioning, brand strategy, and channel economics all bear on the final pricing decision. Present the full demand curve to stakeholders so that the trade-offs at each price point are visible, rather than a single number that obscures context.

Frequently asked questions

What is Gabor-Granger pricing?

Gabor-Granger pricing is a survey-based method for measuring price elasticity and willingness to pay. Respondents are asked whether they would purchase a product at a specific price. By varying that price across respondents or rounds, researchers build a demand curve and identify the price that maximises expected revenue.

How many price points should a Gabor-Granger study include?

Between 5 and 15 price points is the typical range. Most studies use 6 to 10 evenly or strategically spaced prices to balance demand curve resolution against respondent fatigue.

What sample size do I need?

A minimum of 100 completed responses per segment is the standard rule. If you need to compare sub-groups, such as heavy versus light users or different geographies, plan for at least 100 responses per sub-group.

What is the difference between Gabor-Granger and Van Westendorp?

Van Westendorp asks four open-ended questions to identify a range of acceptable prices, without requiring the researcher to specify prices in advance. Gabor-Granger tests specific predefined prices to find the revenue-optimal point within a range. The two methods are complementary. Van Westendorp is better for early-stage exploration. Gabor-Granger is better once a plausible range is established.

Conclusion

Gabor-Granger pricing research has been a fixture of brand and pricing strategy for over sixty years because it produces outputs that are directly actionable. A demand curve, a revenue curve, and a clear revenue-optimising price point give commercial teams the evidence they need to move from instinct to informed decision. Paired with Van Westendorp for early-stage exploration, or with conjoint analysis when product attributes are also in play, it sits at the core of any serious willingness to pay research programme.

The challenge, as with all survey-based pricing methods, is the gap between what people say they will do and what they actually do when their own money is on the line. Minimising that gap requires rigorous sampling, careful questionnaire design, and honest calibration of stated intent. For teams that want to pressure-test a price range before committing budget, an AI panel modelled on your audience and benchmarked against a human panel lets you run Gabor-Granger analysis on synthetic respondents first, so you can refine your price range and question design before a single real respondent is recruited.

To understand how an AI panel complements traditional pricing surveys, read our guide on testing your audience assumptions.